Viral infections are a major threat to human health and account for many serious infectious diseases. Hepatitis C virus (HCV), a major cause of viral hepatitis, infects more than 200 million people worldwide. Current treatment for HCV infection is restricted to immunotherapy with interferon-α, alone or in combination with Ribavirin, a nucleoside analog. This treatment is effective in only about half of the patients. Hepatitis B virus (HBV) acutely infects almost a third of the world's human population, and about 5% of the infected are chronic carriers of the virus. Chronic HBV infection causes liver damage that frequently progresses to cirrhosis and liver cancer later in life. Despite the availability and widespread use of effective chemotherapy and vaccines, the number of chronic carriers approaches 400 million worldwide.
Human immunodeficiency virus (HIV) causes progressive degeneration of the immune system, leading to the development of AIDS. A number of drugs have been used clinically, including HIV nucleoside and non-nucleoside reverse transcriptase inhibitors and protease inhibitors. Currently, combination therapies are the accepted standard of treatment of AIDS in order to reduce viral load and the emergence of drug resistance. Despite progress in the development of anti-HIV drugs, AIDS is still one of the leading epidemic diseases. Therefore, there is still an urgent need for new, more effective HCV, HBV, and HIV drugs. The treatment of viral infections caused by other viruses such as HSV, CMV, influenza viruses, West Nile virus, small pox, EBV, VZV, and RSV also needs better medicines.
Bacterial infections long have been the sources of many infectious diseases. The widespread use of antibiotics results in the emergence of drug-resistant life-threatening bacteria. Fungal infections are another type of infectious disease, some of which also can be life-threatening. There is an increasing demand for the treatment of bacterial and fungal infections. Antimicrobial drugs based on new mechanisms of action and new chemical classes are especially important.
Cellular proliferative disorders are responsible for numerous diseases resulting in major morbidity and mortality and have been intensively investigated for decades. Cancer now is the second leading cause of death in the United States, and over 500,000 people die annually from this proliferative disorder. All of the various cells types of the body can be transformed into benign or malignant tumor cells. Transformation of normal cells into cancer cells is a complex process and, thus far, is not fully understood. Treatment of cancer normally consists of surgery, radiation, and chemotherapy. While chemotherapy can be used to treat all types of cancer, surgery and radiation therapy are limited to certain cancers at various stages of growth at certain sites of the body. There are a number of anticancer drugs widely used clinically. Among them are alkylating agents such, as cisplatin, and antimetabolites, such as 5-Fluorouracil, Gemcitabine, Cytarabine, Fludarabine, and Cladarabine. Although surgery, radiation, and chemotherapies are available to treat cancer patients, there is no cure for cancer at the present time. Cancer research is still one of the most important endeavors in medical and pharmaceutical organizations.
Nucleoside analogs have been used clinically for the treatment of viral infections and proliferative disorders for decades. Most of the nucleoside drugs are classified as antimetabolites. After they enter cells, nucleoside analogs are successively phosphorylated to nucleoside 5′-monophosphates, 5′-diphosphates, and 5′-triphosphates. In most cases, nucleoside triphosphates, e.g. 3′-azido-3′-deoxythymidine (AZT, an anti-HIV drug) triphosphate and arabinosylcytosine (Cytarabine, an anticancer drug) triphosphate, are the chemical entities that inhibit DNA or RNA synthesis, either through a competitive inhibition of polymerases or through incorporation of modified nucleotides into DNA or RNA sequences. Nucleosides may act also as their diphosphates. For instance, 2′-deoxy-2′,2′-difluorocytidine (Gemcitabine, an anticancer drug) 5′-diphosphate has been shown to inhibit human ribonucleotide reductase. Current clinically-used nucleoside drugs primarily depend on cellular activation by nucleoside kinases (nucleoside to nucleotide) and nucleotide kinases (nucleotides to di- and tri-nucleotides). Efficient phosphorylation at each step is required for a nucleoside to be an effective drug.
In order to overcome the usual deficiencies of cellular phosphorylation of unnatural nucleosides, nucleotides, themselves, have been considered as antimetabolite drugs. However, the multiply-charged nucleotides do not effectively penetrate cell membranes and often are hydrolyzed by certain extracellular enzymes. In the last two decades, nucleoside mono-phosphate prodrugs have been intensively investigated as an alternative drug form (Wagner et al., Med. Res. Rev. 2000, 20, 417-451; Jones et al., Antiviral Res. 1995, 27, 1-17; Perigaud et al. Adv. in Antiviral Drug Des. 1995, 2, 147-172; Huynh-Dinh, Curr. Opin. Invest. Drugs, 1993, 2, 905-915). It was hoped that nucleoside mono-phosphate prodrugs, which mask the negative charges on the phosphate by reversible chemical modifications, now being much more lipophilic, would transverse cell membranes and liberate the nucleoside mono-phosphate intracellularly. Cleavage of the prodrug moiety from the nucleoside mono-phosphate would proceed enzymatically via a variety of ubiquitous, non-specific enzymes, like esterases or hydrolytically. Having now bypassed the first kinase step, which is often the most difficult of the three steps with unnatural nucleoside, higher concentration of the required, active species, the nucleoside tri-phosphate was expected. Progress in the area of nucleotide phosphate prodrugs has been made. For instance, certain phosphate prodrugs of anti-HIV nucleosides have been explored for their use as antiviral drugs. The di- and tri-phosphates of 3′-deoxy-3′-azidothymidine (AZT) and 2′,3′-didehydro-2′,3′-dideoxythymidine (D4T) were converted to their acyl prodrugs (Bonnaffe et al. J. Org. Chem. 1996, 61, 895-902). AZT di- and tri-phosphate prodrugs demonstrated similar inhibition of HIV-infected cells as AZT itself, while the corresponding D4T di- and tri-phosphate prodrugs exhibited lower, but still significant anti-HIV activity. Since the acylphosphate moiety of the prodrugs is sensitive to chemical hydrolysis, it is assumed that the prodrugs had been converted to AZT and D4T before they enter cells. Phospholipids also have been as the masking moiety of nucleoside mono- and di-phosphates. AZT di-phosphate tethered with a thioether lipid showed potent inhibition of HIV-infected CEM cells (Hong et al. J. Med. Chem. 1996, 39, 1771-1777). Other lipid-tethered nucleoside di- and tri-phosphates also have been studied (Hostetler et al., J. Biol. Chem. 1990, 265, 6112-6117). Some antitumor nucleosides were also converted to the corresponding nucleotide prodrugs aimed at enhancing antitumor activities. Treatment with lipid-tethered Ara-C di-phosphates demonstrated longer life-span of p388-infected mice than that with Ara-C itself (Hong et al., J. Med. Chem. 1986, 29, 2038-2044; 1990, 33, 1380-1386). 8-Aza-2-deoxyadenosine and 8-bromo-2-deoxyadenosine, two weakly cytotoxic agents, were converted to their 5′-bis(pivaloxymethyl)phosphate prodrugs, which exhibited significantly improved cytotoxicity (Rose et al., J. Med. Chem. 2002, published on web).
Although the prodrugs of nucleotides bearing natural phosphates exhibited certain in vitro and in vivo activities, several major obstacles remain to be overcome. The most obvious barrier is the inherent instability of the natural phosphates to cellular enzymes. Nucleotide prodrugs may, in certain cases, deliver negatively-charged nucleotides into cells better than the parent nucleotides, but are not significantly stabile towards enzymatic and hydrolytic degradation. In addition, nucleoside phosphates bearing natural phosphates when released from their prodrugs intracellularly, may not be anabolized to the required active species (nucleoside di- or tri-phosphates), but may be catabolized back to the inactive parent nucleoside, which is resistant to phosphoylation. In several cases, not only is the active species not formed in sufficient concentrations to elicit effective therapeutic effects, but instead, an intermediate nucleoside phosphate that is formed may be a toxic species. As a case in point, AZT mono-phosphate accumulates in cells because the nucleoside mono-phosphate is a poor substrate for thymidylate kinase and is thought to be responsible for cellular toxicity.
In order to stabilize nucleotides, several nucleoside phosphates bearing di-phosphate or tri-phosphate mimics have been prepared and some of them have been evaluated various biological assays. Many nucleotide mimics or their biological use have been disclosed (Eckstein et al. U.S. Pat. No. 3,846,402 issued November 1974; Horwitz et al., U.S. Pat. No. 4,266,048 issued May 1981; Schinazi et al., U.S. Pat. No. 5,118,672 issued June 1992; Ingels et al., U.S. Pat. No. 5,721,219 February 1998; Bottaro, et al., U.S. Pat. No. 6,303,774 October 2001; Boucher et al. U.S. Pat. No. 6,143,279 issued November 2000; Johansson, EP0357571, 7/1990; Lebeau et al. WO9600733, 1/1996; Vladimirovich et al., WO9820017, 5/1998; Watanabe, WO0179246, 4/2001; Yerxa et al., WO0145691, 6/2001; Peterson, WO0187913, 11/2001). The early work in the chemistry and biological evaluations of nucleotide mimics have been reviewed (Scheit, K. H., Nucleotide Analogs, John Wiley & Sons, New York, (1980); Engel, R., Chem. Rev. 1977, 77, 349-467; Yount, R. G., Adv. in Enzymol. 1975, 43, 1-56).
One type of nucleoside di- and tri-phosphate mimic has modifications at the bridging positions of nucleoside diphosphates and triphosphates (Yount et al., Biochemistry 1971, 10, 2484-2489; Ma et al., J. Med. Chem. 1992, 35, 1938-1941; Ma et al., Bioorg. Chem. 1989, 17, 194-206; Li et al., Bioorg. Chem. 1996, 24, 251-261; Trowbridge et al., J. Am. Chem. Soc. 1972, 94, 3816-3824; Stock, J. Org. Chem. 1979, 44, 3997-4000; Blackburn et al., J. Chem. Soc. Chem. Comm. 1981, 1188-1190; Shipitsin et al., J. Chem. Soc. Perkin Trans 1, 1999, 1039-1050; Arabshahi et al, Biochemistry, 1990, 29, 6820-6826; Yanachkov et al., Nucleosides Nucleotides 1994, 13, 339-350). Among these phosphate mimics are the β,γ-imidotriphosphates, β,γ-methylimidotriphosphates, α,β-imidotriphosphates, α,β:β,γ-diimidotriphosphates, α,β-methylenetriphosphates, β,γ-methylenetriphosphates, α,β:β,γ-bismethylenetriphosphates, α,β-dihalomethylenetriphosphates, α,β-dihalomethylenetriphosphates, β,γ-halomethylenetriphosphates, and α,β-halomethylenetriphosphates. These phosphate mimics usually enhance the stability of the nucleotide towards hydrolysis by cellular enzymes. Methylene and halomethylenes render the nucleoside di- and tri-phosphate mimics considerable more stable to both chemical and enzymatic hydrolysis.
Another type of nucleoside di-phosphate and tri-phosphate mimic is the substitution of one or more phosphate non-bridging oxygen with other heteroatoms or functional group (Ludwig et al., J. Org. Chem. 1991, 56, 1777-1783; Dineva, Nucleosides Nucleotides 1996, 15, 1459-1467; Dyatkina et al. Nucleosides Nucleotides 1995, 14, 91-103; He et al., J. Org. Chem. 1998, 63, 5769-5773; He et al, Nucleic Acids Res. 1999, 27, 1788-1794; Meyer et al., EMBO 2000, 19, 3520-3529; Arzumanov et al., J. boil Chem. 1996, 271, 24389-24394). Among these phosphate mimics are α-O-alkyltriphosphate, α-O-aryltriphosphate, α-P-alkyltriphosphate, γ-P-aryltriphosphate, α-P-alkylaminotriphosphate, α-P-thiotriphosphate, α-P-boranotriphosphate, γ-O-alkyltriphosphate, γ-O-aryltriphosphate, γ-P-alkyltriphosphate, γ-P-aryltriphosphate, γ-P-alkylaminotriphosphate, γ-P-thiotriphosphate. This type of modification on α- or β-phosphorus usually produces diastereomers due to the formation of a chiral phosphorus center. These nucleoside phosphate mimics generally are more stable to cellular nucleases than natural nucleoside phosphates.
Other nucleoside di- and tri-phosphate mimics include modifications at the 5′-position of nucleosides. For instance, 3′-azido′-3′,5′-dideoxy-5′-methylenethymidine 5′-C-triphosphate in which the 5′-oxygen is replaced with methylene was synthesized and evaluated for anti-HIV activity (Freeman et al., J. Med. Chem. 1992, 35, 3192-3196). The nucleotide mimics in which the 5′-oxygen is replaced by sulfur or amino also were reported (Trowbridge et al., J. Am. Chem. Soc. 1972, 94, 3816-3824; Letsinger et al., J. Am. Chem. Soc. 1972, 94, 292-293; Scheit et al., J. Carbohydr. Nucleosides Nucleotides 1974, 1, 485-490). There are very few nucleotide mimics comprising combinations of two or more phosphate modifications. So far, only nucleotide mimics containing α,β:β,γ-diimidotriphosphate, α,β:β,γ-bismethylenetriphosphate, and α-P-borano-α-P-thiotriphosphate were reported, which contain two modifications each. The parent nucleosides for the preparation of nucleotide mimics in the reported work are generally selected from natural nucleosides and a few well known antiviral nucleosides such as AZT, D4T, and 3′-deoxythymidine.
Some of these nucleotide mimics have been evaluated for their biological activity. AZT 5′-α-P-boranotriphosphate and D4T 5′-α-P-boranotriphosphate exhibited very potent inhibition of HIV reverse transcriptase (RT) with Ki values in the low nM range in assays using homopolymer templates. AZT 5′-β,γ-difluoromethylenetriphosphate and AZT 5′-β,γ-imidotriphosphates also exhibited significant inhibition of DNA polymerase or HIV RT. The negatively-charged nucleotide mimics are not likely to be taken up intact by cells, and no meaningful cell-based antiviral data for di- and tri-nucleotide mimics has ever been published. These nucleotide mimics that are active in cell free biochemical assays contain only one modification each, either at the triphosphate bridging position or simply as a substitution of a phosphate oxygen and, thus, are ready substrates for enzymatic hydrolysis. Therefore, the mimics rapidly are degraded extracellularly to provide the parent nucleoside. Any biological activity would result from the parent nucleoside being taken into cells and anabolized to an active nucleotide.
Nucleotide di- and tri-phosphate mimics that are resistant to cellular enzymes and demonstrate significant biological activities have not been disclosed. The several known nucleotide mimics are constructed from natural nucleosides, such as adenosine, or from known biologically active nucleosides, such as AZT. Furthermore, it is essential that novel nucleoside di- and tri-phosphate mimics that are resistant to enzymatic degradation possess one or more prodrugs to allow effective intracellular transport. Nucleoside di- and tri-phosphate mimics with attached prodrugs have not been disclosed in the literature.